Neural Networks for Pattern Recognition
Neural Networks for Pattern Recognition
Whitening-based feature space transformations in a speech impediment therapy system
TSD'07 Proceedings of the 10th international conference on Text, speech and dialogue
| Hi-index | 0.00 |
The hearing impaired have always had difficulties learning to speak because their auditory feedback is either damaged or missing. The SpeechMaster software package provides real-time visual feedback as a substitute for this. Within the package the forms of the feedback are clear and simple. For instance in the first phase of vowel learning the software uses an effective phoneme recognizer providing real-time visual feedback. In this case flickering letters indicate correctness where the brightness of the letters is proportional to the output of speech recognizer. These unambiguous solutions help the hearing impaired to learn the correct association between the phoneme – grapheme pairs or the connection between their own articulation and the speech signal they produce. Also, with the aid of the computer, children can practice without the need for the continuous presence of the teacher. This is a significant step in the education of the hearing impaired as their traditional therapy includes a long and tedious fixation phase. Furthermore, the use of computer exercises, which are popular with children, speeds up the learning process